Processor Scheduling

Question 1

How is First Come First Served Scheduling Implemented?

Answer 1

Allocates processors base don creation time of process. All processes are executed from start to finish.

Question 2

What are the Pros of First in First out?

Answer 2

Minimises process switches, every process will eventually get processor time

Question 3

What are the cons of First in First out?

Answer 3

Average waiting time is long

Question 4

How is Round Robin Scheduling Implemented?

Answer 4

First Come First Serve but with a limit for given to how long a process can have in the processor in one go (Time Quantum). When this has been reached it goes to the back of the queue until it comes back around

Question 5

How is priority implemented in Round Robin Scheduling?

Answer 5

New processes gain priority in the Queue over processes that have previously had time on the processor.

Question 6

What are the Pros of Round Robin Scheduling?

Answer 6

Distributes Resources in a “fair” manner, one process does not take all the CPU time.
A quick process can pass relatively quickly.

Question 7

What are the Cons of round Robin Scheduling?

Answer 7

Long Average waiting time when processes require multiple time quanta. Performance depends heavily on length of time Quantum.

Question 8

How is Shortest Process Next Scheduling Implemented?

Answer 8

Shares the processor based on process execution time. Shortest goes first, slowest goes last.

Question 9

What are the two types of Shortest Process Next Scheduling?

Answer 9

Preemptive and Nonpreemptive

Question 10

What is the different between preemptive and Nonpreemeptive scheduling?

Answer 10

Preemptive interrupts processes while they are being executed in case of a shorted process starting. Nonpreemeptive doesn’t allow the interruption of current process.

Question 11

What are the Pros of Shortest Process Next Scheduling?

Answer 11

Gives minimum average waiting time for a given set of processes assuming all execution times are estimated prefectly.

Question 12

What are the Cons of Shortest Process Next Scheduling?

Answer 12

Execution time has to be estimated, estimates are not going to be perfect. Longer processes may have waited a long time if there is a constant influx of short processes.

Question 13

What is multilevel queueing?

Answer 13

Different types of processes are queued in their own queue and the appropriate algorithm is used for each queue.

Question 14

What three types are processes and the queues split into?

Answer 14

Interactive processes (interact with I/O), normal processes (system services), Batch processes (processes with I/O devices, pure number crunching)

Question 15

How is priority split between queues?

Answer 15

Fixed Priority scheduling, time slicing

Question 16

What are the Pros of Multilevel queueing?

Answer 16

It is complex and can accommodate a range of different performance objectives.

Question 17

What are the Cons of Multilevel queueing?

Answer 17

Because it is complex it can be difficult to calibrate and lots of decisions need to be made. Some processes can change their characteristics over time.

Question 18

Which two metrics might be used to choose an an algorithm?

Answer 18

User oriented: Focus on performance as perceived by the user
System Oriented: Focus on efficient utilisation of processor.

Question 19

What type of data is stored in Cache?

Answer 19

The most frequently used data, or the data being used at this moment in time in the process.

Question 20

What are the two types of queues used for Multicore Systems?

Answer 20

Common ready queue, Private queues

Question 21

How is Common ready queue used for multi core systems?

Answer 21

All cores share one large queue, when one core becomes available it take the process at first in the queue and starts executing.

Question 22

How are private queues used in multi core systems?

Answer 22

Each queue has its own specific queue, and each process is assigned to one queue.

Question 23

What two techniques are used to improve performance of multi core processors?

Answer 23

Load Balancing and processor affinity

Question 24

What is load Balancing?

Answer 24

Load balancing attempts to balance processes equally between cores, and is enforced automatically by a common ready queue. You don’t want one core to be in down time when another core has processes waiting for it.

Question 25

What is processor affinity?

Answer 25

Processor Affinity trues to keep the same process on the same core for as long as possible to keep the cache with relevant data. In hard affinity there is no moving whatsoever but in soft affinity processes may move if there is sever processor imbalance.